Absolute calibration

[raacampbell]

Need to make calibration files which contain an image derived from a single-mode fibre at a known distance. We need a way of implementing that. We can then use this to get an absolute measure of wavefront.

[raacampbell]

Add dws.fibrephase which calculates the phase change over the pupil created by a point source a known distance away. 027a59a

[raacampbell]

To confirm this works: Mount camera on a translation stage. Obtain two measurements a known small distance apart (e.g. 200μm apart) and measure the phase change. Check that dws.fibrephase returns the same answer. If it's off by a small amount, can this be explained by the sensor/diffuser distance being a little off? If so, correct it and try again.

If all looks good, then it should be possible to use the above rig to derive a precise value for the sensor/diffuser distance.

Once this is all nailed, then it should be possible to use a reference image that is associated with a known wavefront curvature.

[raacampbell]

Not analysed carefully, but looks good so far. A good analysis would be to fit the recorded wavefronts and use that to estimate distance moved by the camera.

[raacampbell]

Based on theoretical wavefronts we can determine distance sensor has moved to great accuracy: 9ddb5f6

[raacampbell]

The real differences in curvature are similar, but I've not turned these into distances from the sensor yet.